The Model Index
Models/DeepSeek

DeepSeek-V4.1-Flash

DeepSeekOpen weights

DeepSeek-V4.1-Flash is an open-weights AI model from DeepSeek with an index score of 163 (GPT-4 = 100), rank 39 of 217.

163147–179Index score#39 of 217 models
$0.051/ $0.57Price per million tokensinput / output · via OpenRouter
1.05MContext windowtokens it reads at once
10 Sep 2026Released2 months ago

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100DeepSeek-V4.1-Flash

Claude Opus 4.6 reached this score 7 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
90%
Humanity's Last Exam
no result
Mathsevidence
FrontierMath (tiers 1-3)
67%
Codingevidence
SWE-bench Verified
no result
Terminal-Bench 4.0
20%
Agentsevidence
APEX-Agents
40%
Professional workevidence
Finance Agent (v2)
53%
Novel problemsevidence
ARC-AGI-2
73%
Long tasksno evidence yet
METR time horizon
no result

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What others sayrank on 2 leaderboards

The Model Indexthird-party tests, our method#39of 217 · 163
Epoch Capabilities Indexoverall capability from many benchmarks#34of 231 · 157

Best attop three among current models

Not in the top three among current models on any category we track.

Where to use itprices checked 18 Nov 2026

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released10 Sep 2026
Release entryDeepSeek-V4.1-Flash
Scored asDeepSeek V4.1 Flash
Output speed–

DeepSeek releasesaround this one

This page as a chartDownload PNG